0% found this document useful (0 votes)
47 views40 pages

Part 2 - PHM122s

Semiconductors Physics

Uploaded by

motorky500
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
47 views40 pages

Part 2 - PHM122s

Semiconductors Physics

Uploaded by

motorky500
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 40

10/14/2021 Semiconductors 1

OUTLINE
 Carrier Drift
 Band Bending
 Carrier Diffusion
 Total Current Density
 Einstein Relations
 Excess Carriers in Semiconductors
 Carrier injection
 Generation and Recombination
 Continuity Equation
 Poisson and Laplace equations

10/14/2021 Semiconductors 2
Carrier Drift
In this section we would now like to consider the nature of carrier DYNAMICS
under a variety of conditions
Thermal velocity, Drift velocity and Mobility =0

• At room temperature and in the absence of an applied electric


field ( = 0) electrons and holes in a semiconductor undergo a
rapid but random thermal motion (thermal velocity, vth) that
the charge carriers execute due to their thermal energy Schematic path of an electron
in a semiconductor
1 2 3 l
From Thermodynamics mvth  kT & vth 
2 2 tr
where l  mean free distance and
𝑡𝑟  mean free time or relaxation time
In spite of this rapid motion the net current flowing through the crystal is zero
x

• If an electric field is applied to semiconductor, it


superimposes a slow net drift velocity (vd) on top of
the thermal velocity. [Note vd   ]
Combined motion due to random thermal
motion and an applied electric field.

10/14/2021 Semiconductors 3
Carrier Drift
• At steady state, the mean momentum gain should equal the impulse (force x time)
applied to the electron during a mean free time period:

mn*vdn  q x trn


 q x trn qtrn
The electron drift velocity is vdn  *
   n x (1)  n  (2)
mn mn*
q x trp qtrp
For holes: vdp  *
  p x (3)  p  *
(4)
mp m p

 Eqs.(1) &(3) predict a linear dependence of the drift


velocity on electric field which is indeed found at low
electric fields (see measured data at next figure)

 At higher fields however the electron and hole drift


velocities saturate at comparable values to the
thermal velocity of the carriers

 in this regime added energy imparted by the field is


transferred to the lattice rather than increasing the
carrier velocity

10/14/2021 Semiconductors 4
Carrier Drift
Example
For simplicity consider the case of FREE electrons at room temperature (300 K). If
the relaxation time of the electrons at this temperature is 10-12 s, estimate their
DRIFT and THERMAL velocities in the presence of an applied electric field of 10
Vcm-1.

Since the electric field is low, we are in the LINEAR regime and we may estimate
the drift velocity as

qtrn 1.6  1019  1012 3


vd    31
10  176 ms 1
m 9.1 10

To estimate the thermal velocity we recall that in a CLASSICAL gas the AVERAGE
thermal energy associated with each particle is 3kT/2
3kT 3  1.38  1023  300
 vth   31
 117 kms 1
m 9.1  10

10/14/2021 Semiconductors 5
Carrier Drift
• Eq.(1) states that the electron drift velocity is proportional to the applied electric field.
The proportionality factor is called the electron mobility μn with units of m2/V-s,

Mobility is an important parameter for carrier transport because it describes how


strongly the motion of an electron or a holes is influenced by an applied electric field.

• Mobility depends strongly on Temperature and Doping Concertation.

silicon at 300 K

silicon

The mobility decreases with increasing For a given temperature, the mobility decreases
temperature because of increasing lattice with increasing impurity concentration because
vibration with increasing temperature of enhanced impurity scatterings.

10/14/2021 Semiconductors 6
Mobility and Scattering (Mathiessen's rule)
• The mobility parameter is a measure of the ease of carrier motion in a crystal.
 The carrier mobility varies inversely with the amount of scattering taking
place within the semiconductor.
 The dominant scattering mechanisms are typically (i) lattice scattering involving
collision with thermally agitated lattice atoms, and (ii) ionized impurity

• The probability that a carrier is scattered in a time dt only


e
by lattice scattering is where is the average time
between two lattice scattering events.

• Similarly, the probability that a carrier is scattered only by e


ionized impurity is where is the average time
between two ionized impurity scattering events.

• The total probability that a carrier is scattered in e


the time interval dt is then the sum of the probabilities of
being scattered by each mechanism.
dt dt dt 1 1 1
    
tr t L t I tr t L t I
̅
Because the mobility is related to relaxation time by = ∗ , we can write last Eq. as

1 1 1
 
 L I

In general, each scattering mechanism is associated with a specific mobility.


The net mobility is determined by Mathiessen's rule :

1/ = 1/1 + 1/2 + 1/3 +…………..

The lowest mobility is the dominating one.

Problem set 2, Solve: T&F 1 1 1


Carrier Drift
Drift Current
• The current (I) across an area in the wire is defined quantitatively as the net charge
flowing across the area per unit time {I = dq/dt } .
I

The current in the wire can be expressed in terms of the


 
   Vd  
A
electrons drift velocity (vd), number of free electrons per

    
unit volume (n) and cross-section area A as follows:

 Suppose all electrons moving with an average drift


velocity vd in the opposite direction of the field  vd dt

 In a time dt each electron advances a distance vddt. In this time, the number of electrons
crossing any cross-section of the wire will be contained in a volume Avdt. The number of
electrons in this volume is nAvdt.

 If each electron has a charge e, then the charge flowing across the cross-section during
time dt is
dq  enAv d dt
dq
Hence I   enAv d  vd  
dt
then I  enA Drift Current

10/14/2021 Semiconductors 9
Carrier Drift
• The Current Density (J) is defined as the current per cross-sectional unit area
I
J   env d  en 
A
J   Ohm’s law
where  = en is the conductivity of the material
note that the resistivity  = 1/ =1/ en

• When the electric field is applied to any semiconductor material

 The hole (positive charge) moves in the OPPOSITE Jp


direction to the electron in the presence of an applied h
h
electric field e
e
h
 The direction of motion of the positive carriers has been e
conventionally assumed to be the direction of a current. e h Jn
 The currents carried by the electrons and holes both
point in the same direction as the electric field however 
and so add rather than canceling

10/14/2021 Semiconductors 10
Carrier Drift

Electron current density is: J n  qnvn  qnn   n  (5)

Hole current density is: J p  qnv p  qp  p   p  (6)

where n and p are the electron and hole conductivity of the material respectively.

Total drift current density is: Jt  Jn  J p (amps/m2)

Total conductivity is: t  n  p (ohm-m)-1


 qn n  q  p p

1 1
Total resistivity is:   (ohm-m)
 qn n  q p p

10/14/2021 Semiconductors 11
Band Bending
• In the absence of an applied voltage or electric V
field the potential energy of the charge carriers
is independent of position and the energy bands V 0 V 
are consequently flat dV
  0
dx 0
• When a voltage is applied across a uniformly x

doped semiconductor, the carrier potential Ec
-qV
energy becomes a function of position and Ei
so causes bending of the energy bands. Ec(x)
Ev
Ei (x)
• To determine this band bending we must  Ev(x)
remember that the voltage between two
points measures the work done to move a
Eref
unit positive charge between the points. Eref is any convenient reference energy

 By definition, a positive voltage raises the potential energy of a hole and lowers the
energy of an electron. The band diagram is higher where the voltage is lower.
assume the voltage is dropped
Ec(x) = constant – qV(x) linearly along its length.

The “constant” takes care of the unspecified and inconsequential zero references for Ec and V.

10/14/2021 Semiconductors 12
Band Bending
The “constant” drops out when one considers the electric field.
dV
  
dx In other words, the slope of
Ec, Ev and Ei indicates the
1 dEc 1 dEv 1 dEi
   electric field
q dx q dx q dx

Example

Electron energy
Next figure presents an energy band diagram of non-
uniform doped silicon semiconductor sample at 300 K with
Ei -Ef = EG/4 at x =  L and Ef -Ei = EG/4 at x = 0. Use Ef
0
as the energy reference level. Use the cited energy band
diagram to answer the following questions

• The direction of drift electron current at x= -L/2 is


(a) + x direction (b) - x direction (c) no current

• The potential energy of electron 3 and hole1 are


(a) 0.84 &0.84 (b) 0.84 & 0 (c) 0.84 & 0.28 (d) 0 & 0.28 Electrons 1 and
2 have the same
• The kinetic energy of electron 2 and potential energy of hole 2 are PE
(a) 0.56 & 0 (b) 0 & 0.65 (c) 0.56 & 0.28 (d) 0 & 0.28

10/14/2021 Semiconductors 13
Carrier Diffusion
Diffusion is driven by the existence of carrier concentration gradients in the semiconductor
and the desire to achieve a uniform distribution of carriers
To drive the expression of diffusion current, let us define the flux (j) as amount of
particles passing per unit time per unit area.
Assumptions: Electron density n(x) varies along the

Electron Density n(x)


x direction, no applied electric field and uniform n(l)

temperature
n(0)
At one mean free path away from x = 0 to the left,
half the electrons n(-l) will cross the plane x = 0 n(-l)

during one mean free time period

The average flux across the plane x=0 from left is -l 0 l


x
n  l  l n  l 
j1   vth _ x
2trn 2
n l 
Similarly, the average flux across the plane x=0 from right is j2  vth _ x
2
vth _ x
The net flux from left to right is j  j1  j2 
2
 n  l   n  l  
10/14/2021 Semiconductors 14
Carrier Diffusion
Approximate the densities at ± l by a Taylor series expansion, we get:

vth _ x   dn   dn   dn
j  n  0   l 
  n  0   l    v l
2   dx  
th _ x
dx   dx

dn dn where Dn = vth-x l
The electron diffusion current is J n diff  q j  qvth _ x l  qDn
is the diffusivity
dx dx
DIFFUSION DIFFUSION
CURRENT CURRENT

e h
ELECTRON HOLE
DIFFUSION e e DIFFUSION h h
(flux) (flux)
e e e h h h
e e e e h h h h
e e e e e h h h h h

x x
dn dp
J n diff   qDn (7) J p diff   qD p (8)
dx dx
NOTE THE SIGNS in Eq.(7) & Eq.(8)

10/14/2021 Semiconductors 15
Total Current Density
In the presence of an applied electric field both carrier drift and diffusion are effective,
we may therefore write expressions for the net electron and hole currents as:
dp
J P  J p drift  J p diffusion  q p p  qD p (9)
dx
dn
J n  J n drift  J n diffusion  qn n  qDn (10)
dx

Total current density J  JP  Jn (11)

ASSIGNEMENT
Next figure is a part of the energy band diagram of a p-type
semiconductor bar of length L. The valence band edge is
sloped because doping is non-uniform along the bar. Assume
that Ev rises with a slope of Δ L
(a) Write an expression for the electric field inside this
semiconductor bar.
(b) what is the electron concentration n(x) along the bar?
Assume that n(x = 0) is no. Express your answer in terms of
no, Δ, and L.

10/14/2021 Semiconductors 16
Einstein Relations
1 * 2 3 qtrn
Recall that mn vth  kT & n 
2 2 mn*
The equipartition theorem in statistical mechanics states that the energy is equally divided
in the degrees of freedom. Thus for the one-dimensional case being considered, we have:
1 * 2 1 kT
mn vth _ x  kT vth2 _ x 
2 2 mn*

2 trn kT q  kT 
For electrons Dn  vth _ xl  trn v th  *     n
mn q  q 
Dn kT

n q
Einstein
Dp kT relations
Similarly, for holes 
p q

The Einstein relations show that mobility and the diffusion constant are
proportional to each other at fixed temperature

10/14/2021 Semiconductors 17
Example

Consider a silicon sample at 300 K that is doped with acceptors at a concentration of


1015 cm-3. A stream of minority carriers is injected at x = 0 and the distribution of these
carriers is assumed to be linear, decreasing from 1011 cm-3 at x = 0 to the equilibrium
value at x = W = 10 mm. Determine the diffusion current density of electrons.

ni  1010 cm 3 , p  N A  1015 cm 3
ni2
n   105 cm 3
p
n  1300 cm 2 / Vs from  vs. N A curve
dn dn
 J n diff  qDn  kT n
dx dx Si at 300K
23 105  1011 
 1.38 10  300 1300   4 
 10 10 
NA
  0.54 mA/ cm 2

The motion of electrons is in the positive x direction Causes a diffusion current in the -x direction!

10/14/2021 Semiconductors 18
Example
A semiconductor maintained at 300 K is
characterized by the energy band diagram
shown below. Use the cited energy band
diagram to select the correct answer.

1. The semiconductor is degenerate at


(a) –W <x< W (b) Near x=0
(c) For all x (d) Nowhere

2. The electric field 


sketch is

3. The electrostatic potential


variation inside the
semiconductor is given by

10/14/2021 Semiconductors 19
4. The kinetic energy of the electron shown on the
diagram is
(a) Epeak -Ec(W) (b) Ec(W) –Epeak
(c) Zero (d) EG + Epeak -Ec(W)

5. Do equilibrium conditions prevail?


(a) Yes (b) No
(c) Can't be determined

6. The electron concentration


plot is roughly given by

7. The electron drift current density flowing at x = xi is


(a) Zero (b) n ni [Epeak -Ec(W)]/W (c) -n ni [Epeak -Ec(W)]/W (d) n ND [Epeak - Ec(W)]/W

8. The hole diffusion current density flowing at x = xi is


(a) Zero (b) p ni [Epeak -Ec(W)]/W (c) q Dp ND/W ( d) -q Dp ND/W

10/14/2021 Semiconductors 20
Excess Carriers in Semiconductors

• In thermal equilibrium the relationship pn = ni2 is valid

• If excess carriers are introduced to a semiconductor (carrier injection) so that pn


> ni2, we have a non-equilibrium situation.

• Another non-equilibrium situation is carrier extraction where pn < ni2

• We can inject carriers by using various methods including:

Optical excitation, applying external electric field (biasing), …

• In the case of optical excitation, the semiconductor absorbs the photon energy (hf)
and an electron-hole pair is generated (generation process)

 Number of excess electrons n = Number of excess holes p

• The magnitude of the excess carrier concentration relative to the majority carrier
concentration determines the injection level.

10/14/2021 Semiconductors 21
Carrier Injection
Thermal equilibrium
Consider an n-type silicon with Nd = 1015 cm-3 under thermal equilibrium. The majority and
minority carrier concentrations are
 nno  Nd = 1015 cm-3 & pno = ni2/nno = 105 cm-3 20
10

Notation: the first subscript refers to the semiconductor 1018


pn nn
type and “o“ refers to the thermal equilibrium condition. 10 16
nno nn
Low-level injection 10 14

 n = p << majority carrier concentration or 12 pn


Nd=1015cm-3
10
approximately the doping concentration
1010

If optical excitation introduces n = p =1012 cm-3 108 ni for Si


Minority carrier: pn increases from 105 to 1012 at 300 oK
6
10
pn>> pno but pn << nno  Nd = 1015 pno
104
Majority carrier: n./ nno = 1012/1015 (0.1%)  nn  nno
102
High-level injection Thermal Low-level High-level
 n = p  majority carrier concentration or equilibrium injection injection
approximately the doping concentration
If optical excitation introduces n = p = 1017 cm-3

10/14/2021 Semiconductors 22
Generation and Recombination
• In case of excess carrier injection, the thermal equilibrium condition is disturbed (np  ni2) and
recombination is the process to restore the system to equilibrium (np = ni2).

• Definitions: the generation rate G is the number of electron – hole pairs generated per unit
volume per second
the recombination rate R is the number of electron – hole pairs annihilated
per unit volume per second
Ec Ec

• When generation or recombination processes


occur between conduction band and valance band Ev Ev
(band to band), it is called direct generation or Band to band generation Band to band Recombination
recombination processes. { n = p } Direct generation and recombination processes

• An intermediate energy states are created in the band gap (ET) by impurities or imperfections
such as dislocations and vacancies. The closer is the trap state to mid gap, the more efficient is
the generation and recombination process. Better to
be near to
Ec E c mid Gap
ET or ET
• The intermediate states in the band gap facilitate
indirect generation and recombination and are Ev Ev
called generation-recombination centers, or traps. Indirect generation and recombination processes

10/14/2021 Semiconductors 23
Generation and Recombination
Surface Recombination
• Because of the abrupt discontinuity of the lattice structure
at the surface, many localized energy states (generation- Surface

recombination centers) may be introduced at the surface


region.
• These energy states, called surface states, may greatly Bulk
enhance the recombination rate at the surface region.

Auger recombination
• It is a non-radiative process involving three carriers.
• Direct Auger recombination occurs when an electron and hole
recombine, but instead of producing light, either an electron is
raised higher into the conduction band or a hole is pushed
deeper into the valence band, as shown in Fig.
• The second electron (hole) becomes an energetic and loses its
energy to the lattice by scattering events.
• Usually Auger recombination is important when the carrier
concentration is very high as a result of either high doping or
high injection level.

10/14/2021 Semiconductors 24
Generation and Recombination
Direct Recombination
The rate of direct recombination is found to be proportional to the number of holes
available in the valence band and the number of electrons in the conduction band
For n-type semiconductor R n   nn p n (12)   proportionality constant, f(T)

Ec
• At thermal equilibrium Gth  Rth   nno pno (13)
Gth Rth
• When light is switched on Gn  GL  Gth Ev
The net rate of change of hole concentration Thermal Equilibrium

dpn
 GL  Gth  Rn  GL  U (14)
dt Ec

where U  net recombination rate = Rn - Gth GL Gth Rn


Ev
In steady state dpn / dt = 0 or GL = U (15)
Under Illumination

• When light is switched off GL  0


dpn
The net rate of change of hole concentration  Gth  Rn  U
dt

10/14/2021 Semiconductors 25
Generation and Recombination
Net Recombination Rate (U)
Consider n-type semiconductor
U  R n  G th   n n p n   n no p no    n no   n  p no   p    n no p no
   n no  p  p no  n   p  n 
For direct recombination, n = p  U    nno  pno  p  p

For low-level injection pn   pno   p   nno


pn  pno
U   nno p  p  p  (16)
p
where p=1/nno is minority carrier (hole) lifetime Minority carrier
lifetime is the mean
time for excess
Similarly for electrons in a p-type U is given as minority carrier to
n p  n po disappear or
U   p po n  n  n  (17) recombine with a
n majority carrier
where n=1/ppo is minority carrier (electron) lifetime

10/14/2021 Semiconductors 26
Continuity Equation
• We have seen that carriers can drift under the influence of an electric field, diffuse when
there is a concentration gradient, created and removed by generation and recombination.
• To describe the behavior of the carriers, we must put all of these effects into one equation.
This is the continuity equation, which is a statement of conservation of charges.
To drive a one-dimensional continuity equation, consider
Jp(x) Jp(x+x)
the infinitesimal volume shown in the figure. The overall
rate of hole increase equals to the algebraic sum of:
A x
The number of holes generated in the slab per second is GpAx x x+x

The number of holes recombined in the slab per second is RpAx

The number of holes entering from the left per second is Jp(x)A/q

The number of holes leaving from the right per second is Jp(x + x)A/q

The rate of hole density change in the slab must then be

p J p  x   J p  x  x 
Ax  A   G p  R p  Ax (18) Conservation
t q of charges

10/14/2021 Semiconductors 27
Continuity Equation
J p  x 
By using Taylor Expansion J p  x  x   J p  x   x
x
p J  x  Ax
By substitution in Eq.(18) Ax   p   G p  R p  Ax
t x q

p 1 J p  x 
Holes continuity eq.   G p  R p 
t q x

Similarly, electrons n 1 J n  x  Recall


  G n  Rn  dp  x 
continuity eq. t q x J P  q p p  x    x   qDp
dx
dn  x 
J n  qn n  x    x   qDn
dx
General holes continuity equation can be written as

p   x  p  x  2 p  x 
  p p  x    p  x   Dp  G p  R p  (19)
t x x x 2
Similarly, the general continuity equation for electrons

n   x  n  x  2n  x 
 n n  x   n  x   Dn  G n  R n  (20)
t x x x 2
10/14/2021 Semiconductors 28
Continuity Equation
Special Case: Under Low Level Injection and Direct Recombination

Recall G  R  GL  Gth  R  GL  U

pn  pno 1
Up  & p 
p  nno
n p  n po 1
Un  & n 
n  p po

Minority carriers (holes) continuity equation in n-type semiconductors will be


pn   x  dpn  x   2 pn  x  pn  pno
   p pn  x    p  x   Dp  G  (21)
t x dx x 2
L
p

Similarly, Minority carrier (electrons) continuity equation in p-type semiconductors


n p   x  n p  x  2np  x  n p  n po
 n n p  x   n  x   Dn   G  (22)
t x x x 2
L
n

10/14/2021 Semiconductors 29
Continuity Equation
Application: Illustration
illumination and Transient Response
Consider an n-type sample, that is illuminated with light and in which the electron-hole pairs
are generated uniformly throughout the sample with a generation rate GL.

Form Eq.(21), at steady state ( pn/  t =0), for uniform


illumination ( pn/  x =0), with no electric field ( = 0),
excess hole in the semiconductor equals

p  pn  pno   p GL

At an arbitrary time t=0, the light suddenly is turned off and


the time dependent behavior form Eq. (21) is described by :

pn  p  pno  pn  t   pno   p GL exp   t  p 


 U    n
t   p  solution
   pno  p exp   t  p 
Boundary pn  t  0   pno   p GL
conditions The minority carriers (holes)
pn  t     pno
recombine with majority carriers
(electrons) and decay exponentially
with a time constant τp.

10/14/2021 Semiconductors 30
Continuity Equation
Application: Illumination
illumination and Transient Response
Consider an n-type sample, that is illuminated with light and in which the electron-hole pairs
are generated uniformly throughout the sample with a generation rate GL.

Form Eq.(21), at steady state ( pn/  t =0), for uniform


illumination ( pn/  x =0), with no electric field ( = 0),
excess hole in the semiconductor equals

p  pn  pno   p GL

At an arbitrary time t=0, the light suddenly is turned off and


the time dependent behavior form Eq. (21) is described by :

pn  p  pno  pn  t   pno   p GL exp   t  p 


 U    n
t   p  solution
   pno  p exp   t  p 
Boundary pn  t  0   pno   p GL
conditions The minority carriers (holes)
pn  t     pno
recombine with majority carriers
(electrons) and decay exponentially
with a time constant τp.

10/14/2021 Semiconductors 31
Continuity Equation
APPLICATION: Steady State Injection from One Side - Diffusion Length
Consider an n-type semiconductor and excess carrier is photo-injected from the left end.
From Eq.(21), at steady state under no electric
field and no light inside the semiconductor
 2 pn pn  pno  2 pn pn  pno
Dp  0  0
x 2 p x 2 D p p
The minority carrier diffusion length is defined as
L p  D p p

The minority carriers behavior is described by


 2 pn pn  pno
 0 pn  x   pno   pn  0   pno  e
 x Lp
solution
x 2 2
Lp
Boundary pn  x  0   const .
The minority carrier density decay
exponentially with a characteristic length Lp
conditions p  x     p
n no

Minority carrier diffusion length is the mean distance for a minority carrier to
disappear or recombine with a majority carrier

10/14/2021 Semiconductors 32
Continuity Equation
APPLICATION: Steady State Injection from One Side - Diffusion Length
Consider an n-type semiconductor and excess carrier is photo-injected from the left end.
From Eq.(21), at steady state under no electric
field and no light inside the semiconductor
 2 pn pn  pno  2 pn pn  pno
Dp  0  0
x 2 p x 2 D p p
The minority carrier diffusion length is defined as
L p  D p p

The minority carriers behavior is described by


 2 pn pn  pno
 0 pn  x   pno   pn  0   pno  e
 x Lp
solution
x 2 2
Lp
Boundary pn  x  0   const .
The minority carrier density decay
exponentially with a characteristic length Lp
conditions p  x     p
n no

Minority carrier diffusion length is the mean distance for a minority carrier to
disappear or recombine with a majority carrier

10/14/2021 Semiconductors 33
Poisson and Laplace Equations
Q
Recall Gauss's Law  n dA 
 r
dA
The surface integral of the normal component of the +
Q n
electric field intensity over a closed surface is equal to
1/ times the net charge within that surface
Guassian surface
Poisson’s Equation
Consider an infinitesimal volume with area A and thickness x as shown in the figure.
Let  be the uniform volume charge density, (x) be the normal electric field to the
area A at x and (x+ x ) be the output perpendicular field to area A at (x+ x )
 x A 
From Gauss’s law  ( x   x) A   ( x ) A 
  (x)  (x+ x )
 x
  ( x   x )   ( x) 
 A
x
x x+x
By using Taylor d
Expansion
 ( x   x)   ( x)  x
dx
d  x d 
 ( x)   x   ( x)   (23)
dx  dx 

10/14/2021 Semiconductors 34
Poisson and Laplace Equations
Since the electric field is equal to dV
 
the potential gradient dx
d d 2V
  2
dx dx

d 2V 
From Eq.(23)   Poisson’s Equation
dx 2 
Differential form of Gauss Law
In general for any semiconductor material

  q ( p  n  N D  N A )

Laplace’s Equation
If, as a special case, where volume charge density is zero ( = 0)

d 2V
2
0
dx
10/14/2021 Semiconductors 35
dV 1 dEv 1 
(a)    
dx q dx q L
Ec
Ec(x)
(b) Under thermal equilibrium Ec(0)
n( x )  N c e  ( Ec ( x )  EF )/ kT
also no  N c e  ( Ec (0)  EF )/ kT
Ec is parallel to Ev . Hence, we can write Ec(x) as:

Ec ( x )  x  Ec (0) x
L
So

( x  Ec (0)  EF )/ kT
n( x )  N c e L


 x
 N c e  ( Ec (0) EF )/ kT e LkT


 x
 no e LkT

10/14/2021 Semiconductors 36
 Ec(x)
nn(x)

EF ND

Ei (x) Ev(x)
ni
x
0 x 0

10/14/2021 Semiconductors 37
∆np

∆np(0)

x
-L 0 +L

10/14/2021 Semiconductors 38
np

∆np(0)
∆np(x)
npo

x
-L 0 x +L

10/14/2021 Semiconductors 39
p 1 Jp  x 
  G p  R p
t q x

10/14/2021 Semiconductors 40

You might also like